INDEX

Explanations

not sure

The neuron strongly activates on highly frequent English function words (e.g. “the,” “not,” “often”) and similar common stop-words.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 when

-1.14

？』

-1.13

 ALWAYS

-1.12

 Often

-1.09

著書

-1.07

整體

-1.05

Often

-1.04

ظهار

-1.03

 numerous

-1.03

 some

-1.03

POSITIVE LOGITS

 also

1.19

 predominantly

1.13

只是

1.12

 også

1.10

仅仅

1.09

 however

1.06

 apenas

1.05

只是一

1.05

 instalação

1.05

也很

1.04

Activations Density 0.096%